The main purpose of the central nervous system is to enable animals to generate appropriate behavioral outputs. In order to be appropriate, the animal must take into account both external (sensory) inputs as well as internal (motivational) states. Recent observations suggest that neuronal representations of these internal states are present at relatively early stages in sensory processing. Moreover, these internal states may fundamentally change the way that sensory information is processed by modulating the functional properties of cortical circuits. Yet, the synaptic and circuit mechanisms underlying thi context-dependent processing of sensory information are not known. Moreover, since these changes occur on rapid time scales, standard in vitro methods for investigating the cellular mechanisms of plasticity are not appropriate. Instead, this question requires new approaches for monitoring synaptic connectivity and strength in the awake, behaving animal. Thus, we will develop new strategies, and combine existing ones, to address how cortical circuits are dynamically reconfigured by internal state. In particular, newly developed visually-guided behaviors for headfixed rodents have made the mouse visual system a useful model to study this problem. This proposal builds on these behavioral advances, and combines them with other modern methods including in vivo two-photon calcium imaging, single-cell juxtacellular stimulation, and whole-cell electrophysiology to directly measure how neuronal networks in primary visual cortex are modified during behavior. This novel combination of techniques will reveal how engagement in tasks which rely on the detection of distinct visual features alter the functional connectivity within visual cortex. In addition, we will use newly developed approaches to image the sensory responses in cortico-cortical axonal projections to determine how behavioral context impacts the transmission of sensory information to the higher visual areas. These experiments will identify synaptic and circuit mechanisms that underlie context-dependent changes in network activity, thereby achieving a fundamentally new level of understanding of sensory processing in the actively working brain.